# Face Verification on IJB-A (10 folds average)

Source: [Face Verification on IJB-A (10 folds average)](<https://www.sota2.com/research/sota/face-verification-on-ijb-a-10-folds-average>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-02-26T06:05:50.584Z

Evaluation of face verification performance on the IJB-A dataset, averaged across 10 cross-validation folds, typically measured using True Acceptance Rate (TAR) at various False Acceptance Rates (FAR).

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

| Scope | Recorded value |
| --- | --- |
| Task | Face Verification |
| Dataset (including any stated split) | IJB-A (10 folds average) |
| Metric | TAR @ FAR=0.01 |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 21 of 21 |
| Coverage | All results currently returned for this benchmark |

Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol.

## Metrics

- [TAR @ FAR=0.01](<https://www.sota2.com/research/sota/face-verification-on-ijb-a-10-folds-average.md?metric=TAR%20%40%20FAR%3D0.01>): Higher is better; unit: PERCENT.

- [TAR @ FAR=0.1](<https://www.sota2.com/research/sota/face-verification-on-ijb-a-10-folds-average.md?metric=TAR%20%40%20FAR%3D0.1>): Higher is better; unit: PERCENT.

- [TAR @ FAR=0.001](<https://www.sota2.com/research/sota/face-verification-on-ijb-a-10-folds-average.md?metric=TAR%20%40%20FAR%3D0.001>): Higher is better; unit: PERCENT.

- [FNIR @ FPIR=0.1](<https://www.sota2.com/research/sota/face-verification-on-ijb-a-10-folds-average.md?metric=FNIR%20%40%20FPIR%3D0.1>): Lower is better; unit: PERCENT.

- [FNIR @ FPIR=0.01](<https://www.sota2.com/research/sota/face-verification-on-ijb-a-10-folds-average.md?metric=FNIR%20%40%20FPIR%3D0.01>): Lower is better; unit: PERCENT.

## Recorded results

| Method | TAR @ FAR=0.01 | Result date | Report paper |
| --- | --- | --- | --- |
| Augmented Faces | 88.6 | Not recorded | 68fa62671e209131ce9f92de |
| DCNN | 83.8 | Not recorded | 68fa62621e209131ce9f904c |
| Chen et al. | 83.8 | Not recorded | 68fa62671e209131ce9f92de |
| DCNN | 81.8 | Not recorded | 68fa62621e209131ce9f904c |
| Swami S. et al. | 79 | Not recorded | 68fa62671e209131ce9f92de |
| multi-pose deep face representation | 78.7 | Not recorded | 68fa625c1e209131ce9f8c3b |
| DCNN | 78.7 | Not recorded | 68fa62621e209131ce9f904c |
| Deep Multi-Pose | 78.7 | Not recorded | 68fa62671e209131ce9f92de |
| Proposed Deep Model | 73.3 | Not recorded | 68fa61fe1e209131ce9f6faa |
| Wang et al. | 73.2 | Not recorded | 68fa62621e209131ce9f904c |
| Wang et al. | 73.2 | Not recorded | 68fa62671e209131ce9f92de |
| DCNN | 64 | Not recorded | 68fa62621e209131ce9f904c |
| DCNN | 57.3 | Not recorded | 68fa62621e209131ce9f904c |
| Chen et al. | 57.3 | Not recorded | 68fa62671e209131ce9f92de |
| GOTS | 40.6 | Not recorded | 68fa61fe1e209131ce9f6faa |
| GOTS | 40.6 | Not recorded | 68fa62671e209131ce9f92de |
| OpenBR | 23.6 | Not recorded | 68fa61fe1e209131ce9f6faa |
| OpenBR | 23.6 | Not recorded | 68fa62671e209131ce9f92de |

### Augmented Faces

TAR @ FAR=0.01: 88.6. Result ID: 698c69a837894ef7d9ce3326.

[Reporting paper](<https://www.sota2.com/research/paper/68fa62671e209131ce9f92de.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa62671e209131ce9f92de.md>)

[Implementation](<https://www.sota2.com/www.openu.ac.il/home/hassner/projects/augmented_faces>)

Recorded evaluation configuration: {"Fine-tuning":"false"}. This record does not assert independent reproduction.

### DCNN

TAR @ FAR=0.01: 83.8. Result ID: 698c694c37894ef7d9ce31c0.

[Reporting paper](<https://www.sota2.com/research/paper/68fa62621e209131ce9f904c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa62621e209131ce9f904c.md>)

Recorded evaluation configuration: {"fusion":true}. This record does not assert independent reproduction.

### Chen et al.

TAR @ FAR=0.01: 83.8. Result ID: 698c69a837894ef7d9ce3324.

[Reporting paper](<https://www.sota2.com/research/paper/68fa62671e209131ce9f92de.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa62621e209131ce9f904c.md>)

Recorded evaluation configuration: {"Fine-tuning":"true"}. This record does not assert independent reproduction.

### DCNN

TAR @ FAR=0.01: 81.8. Result ID: 698c694c37894ef7d9ce31bf.

[Reporting paper](<https://www.sota2.com/research/paper/68fa62621e209131ce9f904c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa62621e209131ce9f904c.md>)

Recorded evaluation configuration: {"finetuning":true,"metric learning":true,"color":true}. This record does not assert independent reproduction.

### Swami S. et al.

TAR @ FAR=0.01: 79. Result ID: 698c69a837894ef7d9ce3325.

[Reporting paper](<https://www.sota2.com/research/paper/68fa62671e209131ce9f92de.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa62621e209131ce9f8fc2.md>)

Recorded evaluation configuration: {"Fine-tuning":"true"}. This record does not assert independent reproduction.

### multi-pose deep face representation

TAR @ FAR=0.01: 78.7. Result ID: 698c68dd37894ef7d9ce2ff1.

[Reporting paper](<https://www.sota2.com/research/paper/68fa625c1e209131ce9f8c3b.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa625c1e209131ce9f8c3b.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### DCNN

TAR @ FAR=0.01: 78.7. Result ID: 698c694c37894ef7d9ce31be.

[Reporting paper](<https://www.sota2.com/research/paper/68fa62621e209131ce9f904c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa62621e209131ce9f904c.md>)

Recorded evaluation configuration: {"finetuning":true,"metric learning":true}. This record does not assert independent reproduction.

### Deep Multi-Pose

TAR @ FAR=0.01: 78.7. Result ID: 698c69a837894ef7d9ce3323.

[Reporting paper](<https://www.sota2.com/research/paper/68fa62671e209131ce9f92de.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa625c1e209131ce9f8c3b.md>)

Recorded evaluation configuration: {"Fine-tuning":"false"}. This record does not assert independent reproduction.

### Proposed Deep Model

TAR @ FAR=0.01: 73.3. Result ID: 698bd0e77a967eddb7718e3f.

[Reporting paper](<https://www.sota2.com/research/paper/68fa61fe1e209131ce9f6faa.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa61fe1e209131ce9f6faa.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### Wang et al.

TAR @ FAR=0.01: 73.2. Result ID: 698c694c37894ef7d9ce31bb.

[Reporting paper](<https://www.sota2.com/research/paper/68fa62621e209131ce9f904c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa61fe1e209131ce9f6faa.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### Wang et al.

TAR @ FAR=0.01: 73.2. Result ID: 698c69a837894ef7d9ce3321.

[Reporting paper](<https://www.sota2.com/research/paper/68fa62671e209131ce9f92de.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa61fe1e209131ce9f6faa.md>)

Recorded evaluation configuration: {"Fine-tuning":"false"}. This record does not assert independent reproduction.

### DCNN

TAR @ FAR=0.01: 64. Result ID: 698c694c37894ef7d9ce31bd.

[Reporting paper](<https://www.sota2.com/research/paper/68fa62621e209131ce9f904c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa62621e209131ce9f904c.md>)

Recorded evaluation configuration: {"finetuning":true}. This record does not assert independent reproduction.

### DCNN

TAR @ FAR=0.01: 57.3. Result ID: 698c694c37894ef7d9ce31bc.

[Reporting paper](<https://www.sota2.com/research/paper/68fa62621e209131ce9f904c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa62621e209131ce9f904c.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### Chen et al.

TAR @ FAR=0.01: 57.3. Result ID: 698c69a837894ef7d9ce3322.

[Reporting paper](<https://www.sota2.com/research/paper/68fa62671e209131ce9f92de.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa62621e209131ce9f904c.md>)

Recorded evaluation configuration: {"Fine-tuning":"false"}. This record does not assert independent reproduction.

### GOTS

TAR @ FAR=0.01: 40.6. Result ID: 698bd0e77a967eddb7718e3d.

[Reporting paper](<https://www.sota2.com/research/paper/68fa61fe1e209131ce9f6faa.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### GOTS

TAR @ FAR=0.01: 40.6. Result ID: 698c69a837894ef7d9ce331f.

[Reporting paper](<https://www.sota2.com/research/paper/68fa62671e209131ce9f92de.md>)

Recorded evaluation configuration: {"Fine-tuning":"false"}. This record does not assert independent reproduction.

### OpenBR

TAR @ FAR=0.01: 23.6. Result ID: 698bd0e77a967eddb7718e3e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa61fe1e209131ce9f6faa.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### OpenBR

TAR @ FAR=0.01: 23.6. Result ID: 698c69a837894ef7d9ce3320.

[Reporting paper](<https://www.sota2.com/research/paper/68fa62671e209131ce9f92de.md>)

Recorded evaluation configuration: {"Fine-tuning":"false"}. This record does not assert independent reproduction.

Page 1 of 1. This is an index page, not the full collection.
